Bone marrow stromal cells (BMSCs) contain mesenchymal stem cells that can generate a broad array of tissues including bone. This capacity has numerous clinical applications, especially since alternative approaches to induce new bone formation each have associated limitations. We have demonstrated that BMSCs from both mouse and human sources maintain their osteogenic potential in open systems such as subcutaneous sites in immunocompromised mice where multiple experimental groups can be analyzed in the same animal. This in vivo model of osteogenesis provides a direct test of our hypothesis that the lineage of BMSCs can be directed to a greater percentage of osteoprogenitor cells through genetic and biochemical modifications. In this application, we will develop ex vivo gene therapy strategies to increase the rate, extent and predictability of autogenous BMSC transplants for localized skeletal regeneration. This proposal will focus on several important aspects of skeletal regeneration by bone marrow stromal cells including the ability to direct the lineage of this heterogeneous population of multipotent cells, the determination of an immune response to genetic modification by adenovirus transduction, the enrichment of osteogenic progenitor cells, and the lack of defined criteria for successful bone formation by transplanted BMSCs. Three specific aims are proposed: 1) to determine whether bone-associated transcription factors drive the lineage progression of osteoprogenitor cells in vivo; 2) to elucidate the dependence of proliferative capacity of BMSCs on the osteogenic precursor phenotype and 3) to determine the dependence of osteogenesis in vivo on the expression of phenotypic markers in vitro. By enriching BMSC populations for osteogenic precursors through biochemical and genetic modifications, we will transplant a greater percentage of cells that have the capacity to form new bone in vivo. We will correlate the expression of key phenotypic markers that span the osteoblast lineage from a "potential stem cell" to a mature osteoblast, and thus develop predictor parameters for successful bone formation in vivo. The information gained from this study will provide an essential prerequisite for the development of regenerative approaches that utilize bone marrow stromal cells for gene-and cell-based therapies.